[gmx-developers] Plans for GROMACS 2016 release

Szilárd Páll pall.szilard at gmail.com
Thu Feb 4 14:22:13 CET 2016


Hi,

I know my input on versionining was late (and I guess effectively
invalid), but I'd still like to suggest reconsidering whether that
"20" prefix is really needed; 16 carries the same information that
IIRC the date-based versioning proponents wanted. Or is overflowing in
2099 is a concern? :)

Cheers,
--
Szilárd


On Thu, Feb 4, 2016 at 12:13 PM, Mark Abraham <mark.j.abraham at gmail.com> wrote:
> Hi devs,
>
> I propose a much lower-key release cycle than we have had in the past. It
> has dragged out too long and been too stressful for all involved :-)
>
> On April 1, I'll fool you all and roll a release candidate from master
> branch HEAD, and branch release-2016 immediately. If stuff isn't merged yet,
> then it isn't. The one thing that can't get me fired is releasing quality
> code at a predictable time, and the fewer things we submit to master branch,
> the easier it is for me to do that job! ;-) If you need/want some code in
> the release, then it needs to be in Gerrit with Doxygen, user docs and test
> cases... real soon now! And get your reviewers interested in advance,
> because they're all super busy people. :-)
>
> Then May 1, when the worst of the known issues are handled, I'll release
> GROMACS 2016. That'll then give us ~6 weeks before Swedes disappear on
> summer holidays if we need to do 2016.1 release then.
>
> Our long beta phases for 4.6, 5.0 and 5.1 haven't been terribly effective at
> finding development bugs (e.g. I broke multi-rank-per-simulation replica
> exchange before 5.1, and nobody noticed until the code had been out for
> several months; there's various other such examples). Instead, I'll use that
> month to prioritise the kind of exhaustive testing that has been too onerous
> for me to contemplate doing after a multi-month beta phase. Hopefully some
> of my new testing infrastructure can be reviewed and integrated by then, to
> make that easier.
>
> I am planning to do a GROMACS 2016 paper, and those of you who've been
> working actively and regularly on the wider code base will be most welcome
> to help write that!
>
> Happy coding and reviewing!
>
> Mark
>
> --
> Gromacs Developers mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or
> send a mail to gmx-developers-request at gromacs.org.


More information about the gromacs.org_gmx-developers mailing list